| { | |
| "best_metric": 0.4793087553644425, | |
| "best_model_checkpoint": "trained_models/training_1691463035_ag_news_t5-large/model/checkpoint-24000", | |
| "epoch": 4.0, | |
| "global_step": 24000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 12.1134, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.3869, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1264, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1039, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.0863, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0796, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.0777, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0688, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.072, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0699, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.0666, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.061, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_acc": 0.9465789473684211, | |
| "eval_f1": 0.4774685948550837, | |
| "eval_loss": 0.06575142592191696, | |
| "eval_runtime": 297.0218, | |
| "eval_samples_per_second": 25.587, | |
| "eval_steps_per_second": 3.198, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5.4166666666666664e-05, | |
| "loss": 0.0569, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 5.833333333333334e-05, | |
| "loss": 0.0568, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0566, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0559, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 7.083333333333334e-05, | |
| "loss": 0.0538, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0553, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 7.916666666666666e-05, | |
| "loss": 0.056, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.0524, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.0538, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 9.166666666666667e-05, | |
| "loss": 0.0544, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 9.583333333333334e-05, | |
| "loss": 0.058, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0542, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_acc": 0.9465789473684211, | |
| "eval_f1": 0.47827238398782235, | |
| "eval_loss": 0.061526209115982056, | |
| "eval_runtime": 296.7796, | |
| "eval_samples_per_second": 25.608, | |
| "eval_steps_per_second": 3.201, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.953703703703704e-05, | |
| "loss": 0.0371, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.907407407407407e-05, | |
| "loss": 0.0396, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 9.861111111111112e-05, | |
| "loss": 0.0408, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 9.814814814814815e-05, | |
| "loss": 0.0425, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.768518518518519e-05, | |
| "loss": 0.0429, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 9.722222222222223e-05, | |
| "loss": 0.0419, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 9.675925925925926e-05, | |
| "loss": 0.0411, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 9.62962962962963e-05, | |
| "loss": 0.0429, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 9.583333333333334e-05, | |
| "loss": 0.0402, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 9.537037037037038e-05, | |
| "loss": 0.047, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 9.490740740740742e-05, | |
| "loss": 0.042, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.0383, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_acc": 0.945921052631579, | |
| "eval_f1": 0.47849464202012254, | |
| "eval_loss": 0.05987730622291565, | |
| "eval_runtime": 303.3329, | |
| "eval_samples_per_second": 25.055, | |
| "eval_steps_per_second": 3.132, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 9.398148148148148e-05, | |
| "loss": 0.0237, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 9.351851851851852e-05, | |
| "loss": 0.0226, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 9.305555555555556e-05, | |
| "loss": 0.0262, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 9.25925925925926e-05, | |
| "loss": 0.0268, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 9.212962962962963e-05, | |
| "loss": 0.027, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 9.166666666666667e-05, | |
| "loss": 0.0271, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 9.120370370370371e-05, | |
| "loss": 0.0264, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 9.074074074074075e-05, | |
| "loss": 0.0279, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 9.027777777777779e-05, | |
| "loss": 0.0285, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.981481481481481e-05, | |
| "loss": 0.0314, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 8.935185185185185e-05, | |
| "loss": 0.032, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0302, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_acc": 0.9489473684210527, | |
| "eval_f1": 0.4793087553644425, | |
| "eval_loss": 0.05936088785529137, | |
| "eval_runtime": 301.2156, | |
| "eval_samples_per_second": 25.231, | |
| "eval_steps_per_second": 3.154, | |
| "step": 24000 | |
| } | |
| ], | |
| "max_steps": 120000, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.39508422721536e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |